EEG analysis

imports

library(tidyr)
library(purrr)
library(dplyr)
library(ggplot2)
library(RColorBrewer)
library(reshape2)
library(TTR)
require(smooth)
require(greybox)
require(Mcomp)
library(RColorBrewer)

add_loess <- function(df){
  loess_df <- data.frame(df$timestamps)
  for(i in names(df)){
    if(grepl('absolute', i) & !grepl('loess', i)){
      name <- paste("loess_", i, sep = "")
      print(name)
      df[, name] <- loess(df[,i] ~ df$timestamps, data = df, span=0.65)$fitted
      #loess(value ~ timestamps, data=i, span=0.65)$fitted
      #loess_df$paste("loess_", i) <- loess(value ~ timestamps, data=i, span=0.65)$fitted
    }
  }
  #browser()
  #df <- merge(df, loess_df)
  return(df)
}

load data

inexp_meditation_files = sort(list.files("ffted/",pattern="^0_ffted_med"))
inexp_reference_files = sort(list.files("ffted/",pattern="^0_ffted_ref"))
exp_meditation_files = sort(list.files("ffted/",pattern="^1_ffted_med"))
exp_reference_files = sort(list.files("ffted/",pattern="^1_ffted_ref"))
exp_meditation = list()
for(i in 1:length(exp_meditation_files)) {
  file = exp_meditation_files[i]
  exp_meditation[[i]] <- read.csv(paste("ffted/", file, sep=""))
  exp_meditation[[i]] <- exp_meditation[[i]][rowSums(exp_meditation[[i]] == "-Inf") == 0, , drop = FALSE]
  exp_meditation[[i]] <- add_loess(exp_meditation[[i]])
  print(i)
}

[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 1
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 2
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 3
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 4
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 5
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 6
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 7
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 8

inexp_meditation = list()
for(i in 1:length(inexp_meditation_files)) {
  file = inexp_meditation_files[i]
  inexp_meditation[[i]] <-  read.csv(paste("ffted/", file, sep=""))
  inexp_meditation[[i]] <- inexp_meditation[[i]][rowSums(inexp_meditation[[i]] == "-Inf") == 0, , drop = FALSE]
  inexp_meditation[[i]] <- add_loess(inexp_meditation[[i]])
  print(i)
}

[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 1
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 2
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 3
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 4
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 5
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 6
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 7

exp_reference = list()
for(i in 1:length(exp_reference_files)) {
  file = exp_reference_files[i]
  exp_reference[[i]] <-  read.csv(paste("ffted/", file, sep=""))
  exp_reference[[i]] <- exp_reference[[i]][rowSums(exp_reference[[i]] == "-Inf") == 0, , drop = FALSE]
  exp_reference[[i]] <- add_loess(exp_reference[[i]])
  print(i)
}

[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 1
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 2
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 3
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 4
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 5
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 6
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 7
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 8

inexp_reference = list()
for(i in 1:length(inexp_reference_files)) {
  file = inexp_reference_files[i]
  inexp_reference[[i]] <-  read.csv(paste("ffted/", file, sep=""))
  inexp_reference[[i]] <- inexp_reference[[i]][rowSums(inexp_reference[[i]] == "-Inf") == 0, , drop = FALSE]
  inexp_reference[[i]] <- add_loess(inexp_reference[[i]])
  print(i)
}

[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 1
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 2
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 3
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 4
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 5
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 6
[1] "loess_delta_absolute_1"
[1] "loess_theta_absolute_1"
[1] "loess_alpha_absolute_1"
[1] "loess_beta_absolute_1"
[1] "loess_gamma_absolute_1"
[1] "loess_delta_absolute_2"
[1] "loess_theta_absolute_2"
[1] "loess_alpha_absolute_2"
[1] "loess_beta_absolute_2"
[1] "loess_gamma_absolute_2"
[1] "loess_delta_absolute_3"
[1] "loess_theta_absolute_3"
[1] "loess_alpha_absolute_3"
[1] "loess_beta_absolute_3"
[1] "loess_gamma_absolute_3"
[1] "loess_delta_absolute_4"
[1] "loess_theta_absolute_4"
[1] "loess_alpha_absolute_4"
[1] "loess_beta_absolute_4"
[1] "loess_gamma_absolute_4"
[1] 7

convert time from absolute to relative

convert_time <- function(timestamps){
  time <- timestamps - min(timestamps)
  return(time)
}

subsetting data

subset_and_melt <- function(med, exp, wave, electrodes, melt, melt_all){
  if(med){
    if(exp){
      df <- exp_meditation
    }
    else{
      df <- inexp_meditation
    }
  }
  else{
    if(exp){
      df <- exp_reference
    }
    else{
      df <- inexp_reference
    }
  }
  data = list()
  for(i in 1:length(df)) {
  data[[i]] <- df[[i]][, grepl(paste('(', wave, ')|(', electrodes, ')|(timestamp)', sep = ""), names(df[[i]]))]
  data[[i]]$timestamps <- convert_time(data[[i]]$timestamps)
  if(melt){
    data[[i]] <- melt(data[[i]], id.vars = 'timestamps', variable.name = 'waves')
    data[[i]]$waves <- as.factor(data[[i]]$waves)
  }
  }
  if(melt_all){
    data <- melt(data, id.vars = c('timestamps', 'waves', 'value'))
    data$L1 <- as.factor(data$L1)
  }
  
  return(data)
}

testing

temp_2 <- subset_and_melt(FALSE, TRUE, "beta", "None", TRUE, TRUE)
#ggplot(temp_2[[1]], aes(timestamps,value)) + geom_line(aes(colour = waves))

things to think about: * extreme points, spikes, etc * different smoothing rate * difference between meditative and regular state * difference between experienced meditators and newbies * how to determine the levels and scale it for others * what happens when meditation is stable * explore the coherence between alpha and theta waves

explore experienced/inexperienced distributions of alpha wave

alpha_exp_med = subset_and_melt(TRUE, TRUE, 'alpha', 'ALL', TRUE, TRUE)
alpha_inexp_med = subset_and_melt(TRUE, FALSE, 'alpha', 'ALL', TRUE, TRUE)
alpha_exp_med$exp <- "experienced"
alpha_exp_med$exp <- as.factor(alpha_exp_med$exp)
alpha_inexp_med$exp <- "inexperienced"
alpha_inexp_med$exp <- as.factor(alpha_inexp_med$exp)
temp <- rbind(alpha_exp_med, alpha_inexp_med)
ggplot(temp, aes(value, fill = exp)) + 
  geom_histogram(alpha = 0.5, bins = 100, position = 'identity')

explore experienced/inexperienced distributions of theta wave

theta_exp_med = subset_and_melt(TRUE, TRUE, 'theta', 'ALL', TRUE, TRUE)
theta_inexp_med = subset_and_melt(TRUE, FALSE, 'theta', 'ALL', TRUE, TRUE)
theta_exp_med$exp <- "experienced"
theta_exp_med$exp <- as.factor(theta_exp_med$exp)
theta_inexp_med$exp <- "inexperienced"
theta_inexp_med$exp <- as.factor(theta_inexp_med$exp)
temp <- rbind(theta_exp_med, theta_inexp_med)
ggplot(temp, aes(value, fill = exp)) + 
  geom_histogram(alpha = 0.5, bins = 100, position = 'identity')

explore experienced/inexperienced distributions of beta wave

beta_exp_med = subset_and_melt(TRUE, TRUE, 'beta', 'ALL', TRUE, TRUE)
beta_inexp_med = subset_and_melt(TRUE, FALSE, 'beta', 'ALL', TRUE, TRUE)
beta_exp_med$exp <- "experienced"
beta_exp_med$exp <- as.factor(beta_exp_med$exp)
beta_inexp_med$exp <- "inexperienced"
beta_inexp_med$exp <- as.factor(beta_inexp_med$exp)
temp <- rbind(beta_exp_med, beta_inexp_med)
ggplot(temp, aes(value, fill = exp)) + 
  geom_histogram(alpha = 0.5, bins = 100, position = 'identity')

explore experienced/inexperienced distributions of gamma wave

gamma_exp_med = subset_and_melt(TRUE, TRUE, 'gamma', 'ALL', TRUE, TRUE)
gamma_inexp_med = subset_and_melt(TRUE, FALSE, 'gamma', 'ALL', TRUE, TRUE)
gamma_exp_med$exp <- "experienced"
gamma_exp_med$exp <- as.factor(gamma_exp_med$exp)
gamma_inexp_med$exp <- "inexperienced"
gamma_inexp_med$exp <- as.factor(gamma_inexp_med$exp)
temp <- rbind(gamma_exp_med, gamma_inexp_med)
ggplot(temp, aes(value, fill = exp)) + 
  geom_histogram(alpha = 0.5, bins = 100, position = 'identity')

explore experienced/inexperienced distributions of delta wave

t.test(na.omit(alpha_exp_med$value), na.omit(alpha_inexp_med$value), var.equal = TRUE)

    Two Sample t-test

data:  na.omit(alpha_exp_med$value) and na.omit(alpha_inexp_med$value)
t = 101.66, df = 685930, p-value < 2.2e-16
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 0.2013323 0.2092484
sample estimates:
mean of x mean of y 
0.9737287 0.7684384 

t.test(na.omit(alpha_exp_med$value), na.omit(alpha_inexp_med$value), var.equal = TRUE)

    Two Sample t-test

data:  na.omit(alpha_exp_med$value) and na.omit(alpha_inexp_med$value)
t = 150.31, df = 1371900, p-value < 2.2e-16
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 0.2020223 0.2073607
sample estimates:
mean of x mean of y 
0.9738498 0.7691583 

t_test <- function(wave, med){
  wave_exp <- subset_and_melt(med, TRUE, wave, "All", TRUE, FALSE)
  medians_exp <- list()
  for(i in 1:length(wave_exp)){
    medians_exp[[i]] <- median(wave_exp[[i]]$value)
  }
  wave_inexp <- subset_and_melt(med, FALSE, wave, "All", TRUE, FALSE)
  medians_inexp <- list()
  for(i in 1:length(wave_inexp)){
    medians_inexp[[i]] <- median(wave_inexp[[i]]$value)
  }
  
  medians_exp <- unlist(medians_exp, use.names=FALSE)
  medians_inexp <- unlist(medians_inexp, use.names=FALSE)
  return(print(t.test(medians_exp, medians_inexp)))
}

a <- t_test("beta", TRUE)

    Welch Two Sample t-test

data:  medians_exp and medians_inexp
t = -0.18003, df = 7.7255, p-value = 0.8618
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.4753273  0.4068778
sample estimates:
mean of x mean of y 
0.8503494 0.8845742 

a <- t_test("gamma", TRUE)

    Welch Two Sample t-test

data:  medians_exp and medians_inexp
t = -0.43366, df = 7.0362, p-value = 0.6775
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.5995961  0.4135896
sample estimates:
mean of x mean of y 
0.4898110 0.5828143 

a <- t_test("delta", TRUE)

    Welch Two Sample t-test

data:  medians_exp and medians_inexp
t = 0.76505, df = 9.4431, p-value = 0.4629
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.2353533  0.4785046
sample estimates:
mean of x mean of y 
0.3027340 0.1811583 

a <- t_test("theta", TRUE)

    Welch Two Sample t-test

data:  medians_exp and medians_inexp
t = 0.57218, df = 12.632, p-value = 0.5772
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.2152788  0.3697712
sample estimates:
mean of x mean of y 
0.7990211 0.7217749 

a <- t_test("theta", TRUE)

There is no significant statistical difference in medians and means between experienced and inexperienced people

t_test_min <- function(wave, med){
  wave_exp <- subset_and_melt(med, TRUE, wave, "All", TRUE, FALSE)
  medians_exp <- list()
  for(i in 1:length(wave_exp)){
    medians_exp[[i]] <- min(wave_exp[[i]]$value)
  }
  wave_inexp <- subset_and_melt(med, FALSE, wave, "All", TRUE, FALSE)
  medians_inexp <- list()
  for(i in 1:length(wave_inexp)){
    medians_inexp[[i]] <- min(wave_inexp[[i]]$value)
  }
  
  medians_exp <- unlist(medians_exp, use.names=FALSE)
  medians_inexp <- unlist(medians_inexp, use.names=FALSE)
  return(print(t.test(medians_exp, medians_inexp)))
}

a <- t_test_min("alpha", TRUE)
a <- t_test_min("theta", TRUE)
a <- t_test_min("gamma", TRUE)
a <- t_test_min("beta", TRUE)
a <- t_test_min("delta", TRUE)

t_test_max <- function(wave, med){
  wave_exp <- subset_and_melt(med, TRUE, wave, "All", TRUE, FALSE)
  medians_exp <- list()
  for(i in 1:length(wave_exp)){
    medians_exp[[i]] <- max(wave_exp[[i]]$value)
  }
  wave_inexp <- subset_and_melt(med, FALSE, wave, "All", TRUE, FALSE)
  medians_inexp <- list()
  for(i in 1:length(wave_inexp)){
    medians_inexp[[i]] <- max(wave_inexp[[i]]$value)
  }
  
  medians_exp <- unlist(medians_exp, use.names=FALSE)
  medians_inexp <- unlist(medians_inexp, use.names=FALSE)
  return(print(t.test(medians_exp, medians_inexp)))
}

a <- t_test_max("alpha", TRUE)
a <- t_test_max("theta", TRUE)
a <- t_test_max("gamma", TRUE)
a <- t_test_max("beta", TRUE)
a <- t_test_max("delta", TRUE)

#ggplot(temp, aes(value, fill = exp)) + 
#  geom_histogram(alpha = 0.5, bins = 100, position = 'identity')

data_temp <- subset_and_melt(TRUE, TRUE, 'alpha', "ALL", TRUE, TRUE)

#geom_line(aes(y=theta_absolute_2, x = timestamps), 
#       data = without_na, color=brewer.pal(4, "Blues")[3]) + #geom_smooth(aes(y=theta_absolute_2, x = timestamps), 
#       data = without_na, color=brewer.pal(4, "Blues")[3], span = 0.01) +

temp <- es(to_draw$value, h=18, holdout=TRUE, silent=FALSE)

The provided data is not ts object. Only non-seasonal models are available.
Only additive models are allowed with non-positive data.

Forming the pool of models based on... ANN, AAN, Estimation progress:    100%... Done! 

temp <- es(to_draw$value, h=18, holdout=TRUE, silent=FALSE)

#to_draw_part$sma <- sma(to_draw_part$value, n = 5, v = 0.9)$fitted
to_draw_part$loess <- loess(value ~ timestamps, data=to_draw_part, span=0.65)$fitted

ggplot() +
  geom_line(aes(y = value, x = timestamps), data = to_draw_part) +
  geom_smooth(aes(y = value, x = timestamps), data = to_draw_part, span = 1, n = 15, color = "blue") +
  geom_line(aes(y = loess, x = timestamps), data = to_draw_part, color = 'red')

ggplot() +
  geom_smooth(aes(y = value, x = timestamps), data = theta_exp_med[theta_exp_med$waves == 'loess_theta_absolute_2',], span = 1, n = 15, color = "blue") +
  geom_line(aes(y = value, x = timestamps), data = theta_exp_med[theta_exp_med$waves == 'loess_theta_absolute_2',], color = 'red', alpha = 0.2) +
  geom_line(aes(y = value, x = timestamps), data = theta_exp_med[theta_exp_med$waves == 'theta_absolute_2',], color = 'black', alpha = 0.2)

curr_data <- theta_exp_med[theta_exp_med$L1 == '1',]
#[theta_exp_med$waves == 'theta_absolute_2' & 
    
ggplot(aes(y = value, x = timestamps), data = curr_data) +
  geom_line(data = curr_data[curr_data$waves == 'theta_absolute_2', ], color = 'red', alpha = 0.7) +
  geom_line(data = curr_data[curr_data$waves == 'loess_theta_absolute_2',], color = 'black', alpha = 1)

temp <- curr_data[curr_data$waves == 'loess_theta_absolute_2', ]
loess_ordered <- temp$value
loess_ordered <- sort(loess_ordered)
min(loess_ordered)

[1] 0.3447938

barplot(loess_ordered)

temp$state <- cut(temp$value, quantile(temp$value,(0:5)/5))
ggplot(aes(y = value, x = timestamps), data = curr_data) +
  geom_line(data = curr_data[curr_data$waves == 'theta_absolute_2', ], color = 'grey', alpha = 0.7) +
  geom_line(data = curr_data[curr_data$waves == 'loess_theta_absolute_2',], color = 'black', alpha = 1) +
  geom_line(aes(y = value, x = timestamps, color = state, size = 2),data = temp, alpha = 1) 

NA

theta_2, second person

curr_data <- theta_exp_med[theta_exp_med$L1 == '2',]
temp <- curr_data[curr_data$waves == 'loess_theta_absolute_2', ]
temp$state <- cut(temp$value, quantile(temp$value,(0:5)/5))
ggplot(aes(y = value, x = timestamps), data = curr_data) +
  geom_line(data = curr_data[curr_data$waves == 'theta_absolute_2', ], color = 'grey', alpha = 0.7) +
  geom_line(data = curr_data[curr_data$waves == 'loess_theta_absolute_2',], color = 'black', alpha = 1) +
  geom_line(aes(y = value, x = timestamps, color = state, size = 2),data = temp, alpha = 1) 

curr_data <- theta_exp_med[theta_exp_med$L1 == '3',]
temp <- curr_data[curr_data$waves == 'loess_theta_absolute_2', ]
temp$state <- cut(temp$value, quantile(temp$value,(0:5)/5))
ggplot(aes(y = value, x = timestamps), data = curr_data) +
  geom_line(data = curr_data[curr_data$waves == 'theta_absolute_2', ], color = 'grey', alpha = 0.7) +
  geom_line(data = curr_data[curr_data$waves == 'loess_theta_absolute_2',], color = 'black', alpha = 1) +
  geom_line(aes(y = value, x = timestamps, color = state, size = 2),data = temp, alpha = 1) 

curr_data <- theta_exp_med[theta_exp_med$L1 == '4',]
temp <- curr_data[curr_data$waves == 'loess_theta_absolute_2', ]
temp$state <- cut(temp$value, quantile(temp$value,(0:5)/5))
ggplot(aes(y = value, x = timestamps), data = curr_data) +
  geom_line(data = curr_data[curr_data$waves == 'theta_absolute_2', ], color = 'grey', alpha = 0.7) +
  geom_line(data = curr_data[curr_data$waves == 'loess_theta_absolute_2',], color = 'black', alpha = 1) +
  geom_line(aes(y = value, x = timestamps, color = state, size = 2),data = temp, alpha = 1) 

curr_data <- theta_exp_med[theta_exp_med$L1 == '6',]
temp <- curr_data[curr_data$waves == 'loess_theta_absolute_2', ]
temp$state <- cut(temp$value, quantile(temp$value,(0:5)/5))
ggplot(aes(y = value, x = timestamps), data = curr_data) +
  geom_line(data = curr_data[curr_data$waves == 'theta_absolute_2', ], color = 'grey', alpha = 0.7) +
  geom_line(data = curr_data[curr_data$waves == 'loess_theta_absolute_2',], color = 'black', alpha = 1) +
  geom_line(aes(y = value, x = timestamps, color = state, size = 2),data = temp, alpha = 1) 

curr_data <- theta_exp_med[theta_exp_med$L1 == '7',]
temp <- curr_data[curr_data$waves == 'loess_theta_absolute_2', ]
temp$state <- cut(temp$value, quantile(temp$value,(0:5)/5))
ggplot(aes(y = value, x = timestamps), data = curr_data) +
  geom_line(data = curr_data[curr_data$waves == 'theta_absolute_2', ], color = 'grey', alpha = 0.7) +
  geom_line(data = curr_data[curr_data$waves == 'loess_theta_absolute_2',], color = 'black', alpha = 1) +
  geom_line(aes(y = value, x = timestamps, color = state, size = 2),data = temp, alpha = 1) 

curr_data <- theta_exp_med[theta_exp_med$L1 == '8',]
temp <- curr_data[curr_data$waves == 'loess_theta_absolute_2', ]
temp$state <- cut(temp$value, quantile(temp$value,(0:5)/5))
ggplot(aes(y = value, x = timestamps), data = curr_data) +
  geom_line(data = curr_data[curr_data$waves == 'theta_absolute_2', ], color = 'grey', alpha = 0.7) +
  geom_line(data = curr_data[curr_data$waves == 'loess_theta_absolute_2',], color = 'black', alpha = 1) +
  geom_line(aes(y = value, x = timestamps, color = state, size = 2),data = temp, alpha = 1) 

curr_data <- alpha_exp_med[theta_exp_med$L1 == '1',]
temp <- curr_data[curr_data$waves == 'loess_alpha_absolute_2', ]
temp$state <- cut(temp$value, quantile(temp$value,(0:5)/5))
ggplot(aes(y = value, x = timestamps), data = curr_data) +
  geom_line(data = curr_data[curr_data$waves == 'alpha_absolute_2', ], color = 'grey', alpha = 0.7) +
  geom_line(data = curr_data[curr_data$waves == 'loess_alpha_absolute_2',], color = 'black', alpha = 1) +
  geom_line(aes(y = value, x = timestamps, color = state, size = 2),data = temp, alpha = 1) 

alpha_2_all_exp <- subset_and_melt(TRUE, TRUE, 'alpha', 'All', TRUE, TRUE)
ggplot(aes(x = timestamps, y = value), data = alpha_2_all_exp[alpha_2_all_exp$waves == 'loess_alpha_absolute_2',]) +
  geom_line(aes(color = L1)) +
  ggtitle("alpha 2, exp")

alpha_2_all_exp <- subset_and_melt(TRUE, TRUE, 'alpha', 'All', TRUE, TRUE)
ggplot(aes(x = timestamps, y = value), data = alpha_2_all_exp[alpha_2_all_exp$waves == 'loess_alpha_absolute_1',]) +
  geom_line(aes(color = L1)) +
  ggtitle("alpha 1, exp")

alpha_2_all_exp <- subset_and_melt(TRUE, TRUE, 'alpha', 'All', TRUE, TRUE)
ggplot(aes(x = timestamps, y = value), data = alpha_2_all_exp[alpha_2_all_exp$waves == 'loess_alpha_absolute_3',]) +
  geom_line(aes(color = L1)) +
  ggtitle("alpha 3, exp")

alpha_2_all_exp <- subset_and_melt(TRUE, TRUE, 'alpha', 'All', TRUE, TRUE)
ggplot(aes(x = timestamps, y = value), data = alpha_2_all_exp[alpha_2_all_exp$waves == 'loess_alpha_absolute_4',]) +
  geom_line(aes(color = L1)) +
  ggtitle("alpha 4, exp")

alpha_2_all_exp <- subset_and_melt(TRUE, FALSE, 'alpha', 'All', TRUE, TRUE)
ggplot(aes(x = timestamps, y = value), data = alpha_2_all_exp[alpha_2_all_exp$waves == 'loess_alpha_absolute_2',]) +
  geom_line(aes(color = L1)) +
  ggtitle("alpha 2, inexp")

ggplot(aes(x = timestamps, y = value), data = alpha_2_all_exp[alpha_2_all_exp$waves == 'loess_alpha_absolute_1',]) +
  geom_line(aes(color = L1)) +
  ggtitle("alpha 1, inexp")

ggplot(aes(x = timestamps, y = value), data = alpha_2_all_exp[alpha_2_all_exp$waves == 'loess_alpha_absolute_3',]) +
  geom_line(aes(color = L1)) +
  ggtitle("alpha 3, inexp")

ggplot(aes(x = timestamps, y = value), data = alpha_2_all_exp[alpha_2_all_exp$waves == 'loess_alpha_absolute_4',]) +
  geom_line(aes(color = L1)) +
  ggtitle("alpha 4, inexp")

temp <- subset_and_melt(TRUE, TRUE, 'theta', 'All', TRUE, TRUE)
ggplot(aes(x = timestamps, y = value), data = temp[temp$waves == 'loess_theta_absolute_1',]) +
  geom_line(aes(color = L1)) +
  ggtitle("theta 1, exp")

ggplot(aes(x = timestamps, y = value), data = temp[temp$waves == 'loess_theta_absolute_2',]) +
  geom_line(aes(color = L1)) +
  ggtitle("theta 2, exp")

ggplot(aes(x = timestamps, y = value), data = temp[temp$waves == 'loess_theta_absolute_3',]) +
  geom_line(aes(color = L1)) +
  ggtitle("theta 3, exp")

ggplot(aes(x = timestamps, y = value), data = temp[temp$waves == 'loess_theta_absolute_4',]) +
  geom_line(aes(color = L1)) +
  ggtitle("theta 4, exp")

temp <- subset_and_melt(TRUE, TRUE, 'alpha', 'theta', TRUE, TRUE)
ggplot() +
  geom_line(aes(x = timestamps, y = value, color = L1), data = temp[temp$waves == 'loess_theta_absolute_1',],  size = 1) +
  ggtitle("theta 1 and alpha 1, exp") +
  geom_line(aes(x = timestamps, y = value, color = L1), data = temp[temp$waves == 'loess_alpha_absolute_1',], linetype = 'dotted', size = 1) +
  scale_color_brewer(palette = "Spectral")

ggplot() +
  geom_line(aes(x = timestamps, y = value, color = L1), data = temp[temp$waves == 'loess_theta_absolute_2',],  size = 1) +
  geom_line(aes(x = timestamps, y = value, color = L1), data = temp[temp$waves == 'loess_alpha_absolute_2',], linetype = 'dotted', size = 1) +
  scale_color_brewer(palette = "Spectral") +
  ggtitle("theta 2 and alpha 2, exp")

ggplot() +
  geom_line(aes(x = timestamps, y = value, color = L1), data = temp[temp$waves == 'loess_theta_absolute_3',],  size = 1) +
  geom_line(aes(x = timestamps, y = value, color = L1), data = temp[temp$waves == 'loess_alpha_absolute_3',], linetype = 'dotted', size = 1) +
  scale_color_brewer(palette = "Spectral") +
  ggtitle("theta 3 and alpha 3, exp")

ggplot() +
  geom_line(aes(x = timestamps, y = value, color = L1), data = temp[temp$waves == 'loess_theta_absolute_4',],  size = 1) +
  geom_line(aes(x = timestamps, y = value, color = L1), data = temp[temp$waves == 'loess_alpha_absolute_4',], linetype = 'dotted', size = 1) +
  scale_color_brewer(palette = "Spectral") +
  ggtitle("theta 4 and alpha 4, exp")